Niloofar Didar
PROJECTS
My research focus as a PhD candidate is currently on high performance computing, human-computer interaction, and power efficiency for mixed AI-AR apps (Artificial intelligence-Augmented reality). The results of these projects are published or under review in top-conferences/journals. Additionally, I have engaged in a collaborative research project on computational resource management in a shared based edge assisted system with multiple users which resulted in another academic paper currently under review in TPDS journal. During my Internship experience, as well as in my undergraduate, and post-graduate studies, I have been working on many projects in various areas, including edge computing, IoT, web-app development, autonomous vehicles, etc. An overview of each of my projects is provided in the following section.
Latest Projects

Joint AI Task Allocation and Virtual Object Quality Manipulation for Improved MAR App Performance (Dec 2022- Current)- last PhD project
Implementation of a framework called HBO in Android that leverages Bayesian optimization to dynamically trade-off average quality of virtual objects and AI average latency (ML tasks) by utilizing various computing resources such as the GPU, CPU and NPU for on-device inference requests.

Balancing Virtual Objects Quality and AI Throughput in Mobile Augmented Reality Apps (Jan 2021-Dec 2022)-PhD project
Implementation of framework for mobile augmented reality apps in Android to balance the performance of AI and AR tasks on the GPU. It leverages accurate runtime linear model training of performance to manipulate virtual object's total triangle count and employs an approximation algorithm that assigns object triangle count across virtual objects for an enhanced object quality.

Motional Co. Internship Project ( May-Aug 2022)
Design and Implementation of UI/UX and backend of a web-based unified tag management system (using Flask & Java script) for autonomous vehicles and deployment of the Data base using SQLite

Edge Computing - PhD Project (Aug 2021-May 2022)
Enhanced data sharing aware algorithm for task allocation in edge computing systems
DSTAR is a reallocation algorithm that uses a smart rearrangement to allocate some of the unallocated tasks of a previous state-of-the-art algorithm for increased total profit and allocation rate (using Java).

Edge Assisted MAR app- My first PhD Project (Aug 2019- Dec 2020).
eAR: an Edge-Assisted and Energy-Efficient Mobile Augmented Reality Framework
Design an edge-assisted autonomous and energy-efficient framework for MAR apps to maximize their energy efficiency while minimizing the MAR app developer efforts, the total app storage usage, and the impact on the user-perceived quality.

My Master Thesis Project (2016-2018)
IoT-based Intelligent Emergency Response System
This project was to address the immediate fire incident planning based on the idea of providing two phase recommendation system, a plan for evacuating the users, and a plan for rescuing the trapped ones.

My bachelor thesis (2014-2015)
Implementation of the web-based information and communication System for "Computer Science and Engineering Faculty of Shahid Beheshti University"

Internship Project ( May-Aug 2012)
Web-Based Cognitive Testing project which was designed to measure the user's general intelligence. The main goal for the project was to implement a job fit system where a candidate is assessed based on their skills, talent, performance, and willing.